Get started with FlowViz in minutes. Self-host the open-source tool and start analyzing threat reports.
git clone https://github.com/davidljohnson/flowviz.git
cd flowviznpm installCopy the example environment file and configure your LLM API key:
cp .env.example .envThen edit .env and set your API key:
ANTHROPIC_API_KEY=your_anthropic_api_key_hereGet your API key from your LLM provider's console (Anthropic, OpenAI, etc.)
npm run dev:fullOpen http://localhost:5173 in your browser
Paste a URL from any security blog or threat report. FlowViz supports most major security blogs including The Hacker News, Bleeping Computer, Krebs on Security, and more.
AI extracts attack patterns and maps them to MITRE ATT&CK framework. Watch in real-time as the flow builds.
Explore the interactive diagram - zoom, pan, and click nodes for details. Export to PNG, JSON, or STIX format for documentation.
Configure your preferred AI model in the .env file. Models with larger context windows are recommended for complex security analysis.
# Example model configuration
ANTHROPIC_MODEL=claude-sonnet-4-5-20250929
# See .env.example for available optionsModify the AI prompts in src/lib/prompts.ts to customize extraction behavior for your specific use cases.
FlowViz exports STIX 2.1 compatible JSON for integration with threat intelligence platforms. Use the export button after generating a flow.
FlowViz is designed for internal, self-hosted deployment to protect your API keys and data. Follow these secure deployment practices:
npm run build for production